PoMoS & GloMo are two complementary routines developed by Sylvain Mangiarotti, R. Coudret, L. Drapeau & L. Jarlan at CESBIO to obtain a set of differential equations from "measured" time series. The abilities of these algorithms are discussed in [1]
PoMoS (Polynomial Model Search) was written to obain an approximating function combining the successive first n time derivatives to predict the (n+1)th. It is based on an evolutionary algorithm combined to a least square technique.
GloMo (Global Modelling) allows to obtain a low-dimensional global model using a scalar time series [2].
If you are interested by these two routines, you could obtain them by writing to Sylvain Mangiarotti in filling the form available at CESBIO. Please, provide few words at the end of this form about the field in which this source code will be used.
[1] S. Mangiarotti, R. Coudret, L. Drapeau, & L. Jarlan, Polynomial search and global modeling : Two algorithms for modeling chaos, Physical Review E, 86, 046205, 2012. online.
[2] G. Gouesbet & C. Letellier, Global vector field reconstruction by using a multivariate polynomial approximation on nets, Physical Review E, 49 (6), 4955-4972, 1994.